It has become an inevitable trend for greenhouses to develop toward automation and intelligent operations,with the rapid development of agricultural robot technology.Mobile robots have always played an important role in automated picking,handling,spraying,and other tasks in greenhouses.High-precision autonomous navigation achieved is the prerequisite for robots to complete these tasks.Therefore,this paper studies the key technologies of autonomous navigation of mobile robots in the greenhouse environment.Firstly,an omnidirectional motion model based on the Mecanum Wheel was established to realize the trajectory prediction of the mobile robot,to address the problem of narrow and compact road environments in greenhouses.By environmental features of the greenhouse,the navigation scheme of multi-sensor fusion was determined,laying the foundation for the subsequent research.Then,a map construction method based on the fusion of 2D LIDAR and RGB-D camera information was proposed,to address the problems of insufficient environmental information description using alone 2D LIDAR and low accuracy of map construction using RGB-D cameras alone.Mathematical modeling of the sensor was established based on the operating principles of 2D Li DAR and RGB-D cameras.The internal and external parameters of the camera were acquired by Autoware joint calibration method,which was used to construct the transformation matrix of the LIDAR and the camera.The pre-processed camera point cloud data was projected into a two-dimensional pseudo laser point cloud,and the environment grid map was constructed utilizing point cloud superposition and fusion.The experimental results showed that the method effectively compensated for the lack of accuracy of map construction and the lack of the acquired environmental information,improving the reliability and integrity of map construction by mobile robots.Research on the positioning and path planning of mobile robots in the complex environment of greenhouses was studied.The wheel odometer measurement model was used to establish the EKF state equation and the IMU measurement model to establish the EKF observation equation to realize EKF-based fusion positioning.The improved Monte Carlo algorithm was used to further correct the cumulative error of the wheel odometer and IMU.After the positioning was completed,the A* algorithm and the DWA algorithm were used to study the global path planning and local path planning of the robot combined with the characteristics of the greenhouse environment.The experimental results showed that the fusion positioning method effectively improves the positioning accuracy of the robot.And the A* algorithm and the DWA algorithm realized the path planning and real-time obstacle avoidance of the mobile robot.Finally,experiments were conducted to test the accuracy of autonomous navigation of the mobile robot.In the simulated greenhouse environment,by analyzing and comparing the average error of the position and angle of the mobile robot moving to each target point,the effectiveness of the autonomous navigation method of the mobile robot was verified.The results showed that this method can meet the navigation needs of mobile robots in the greenhouse environment. |